skills.rest
eduardtomasek
eduardtomasekCOMMUNITY
Explore all skills by this author →
GitHub

grepai-config-reference

Documents all GrepAI configuration options in .grepai/config.yaml for setup and tuning.

Updated May 12, 2026
Search Tags:#semantic search#embeddings#configuration#code search#yaml#vector store#grepai
One-click install
npx skills add https://github.com/eduardtomasek/codex-dev-stack --skill grepai-config-reference-eduardtomasek
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: grepai-config-reference
Source: https://github.com/eduardtomasek/codex-dev-stack/tree/main/skills/.agents/grepai-config-reference
Command: npx skills add https://github.com/eduardtomasek/codex-dev-stack --skill grepai-config-reference-eduardtomasek

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Configuring GrepAI involves many options across embedders, vector stores, chunking, tracing, and search ranking, and misconfiguration leads to poor indexing or search results. This reference consolidates every available option in one place so you can configure GrepAI correctly the first time. ## Core Features & Use Cases - Complete Schema Reference: Covers embedder providers (Ollama, OpenAI, LM Studio), storage backends (gob, PostgreSQL, Qdrant), chunking, watch, trace, search boosting, and ignore patterns. - Ready-Made Configurations: Provides copy-paste YAML templates for small projects, large codebases, team environments, and maximum-privacy local setups. - Use Case: You are setting up GrepAI on a large monorepo and want semantic search to ignore vendor directories while boosting core source paths. Use this reference to configure the store backend, chunk sizes, and search boost penalties correctly. ## Quick Start Ask the AI to show the recommended GrepAI configuration for a large codebase using a PostgreSQL vector store.

Frequently Asked Questions about grepai-config-reference

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I configure GrepAI for semantic code search?▼

Create a .grepai/config.yaml file in your project root defining an embedder provider, a storage backend, and chunking settings. Run grepai status to validate the configuration before indexing.

Which embedding providers does GrepAI support?▼

GrepAI supports Ollama, OpenAI, and LM Studio as embedder providers. Ollama runs locally with models like nomic-embed-text, while OpenAI offers text-embedding-3-small and text-embedding-3-large.

What storage backends can GrepAI use for vector embeddings?▼

GrepAI supports three storage backends: gob for local file storage, PostgreSQL for scalable team setups, and Qdrant for dedicated vector database deployments with TLS and API key support.

Can I use environment variables in GrepAI config?▼

Yes, GrepAI supports environment variable substitution in config.yaml using the ${VAR_NAME} syntax. This is recommended for API keys and database connection strings so secrets are never committed to the repository.

How do I exclude test files from GrepAI search results?▼

Use the search boost penalties section to reduce scores for paths matching patterns like /tests/ or _test., and add file patterns to the ignore list to exclude them from indexing entirely.

Why is my GrepAI configuration not working?▼

Run grepai status to display configuration errors. Common issues include invalid YAML syntax, unreachable embedder endpoints, and incorrect vector dimensions for the selected embedding model.

Related Skills

Explore More Software Engineering Skills→
jugrajsinghjugrajsingh
1

embedder-config

Manage embedding provider and model configurations for the grepai tool.

Community
Advanced
audibleblinkaudibleblink
4

grepai

Perform semantic code search and trace call relationships across codebases.

Community
Advanced
organvm-i-theoriaorganvm-i-theoria

vector-search-patterns

Implement vector similarity search with embeddings, indexes, and hybrid retrieval pipelines.

Official
Advanced
MadAppGangMadAppGang
278

claudemem-search

Analyze codebases with tree-sitter AST parsing and PageRank symbol ranking.

Official
Advanced